File size: 76,451 Bytes
6fa4bc9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 |
{
"paper_id": "I08-1005",
"header": {
"generated_with": "S2ORC 1.0.0",
"date_generated": "2023-01-19T07:41:06.664311Z"
},
"title": "Semi-Supervised Learning for Relation Extraction",
"authors": [
{
"first": "Guodong",
"middle": [],
"last": "Zhou",
"suffix": "",
"affiliation": {
"laboratory": "Jiangsu Provincial Key Lab for Computer Information Processing Technology",
"institution": "Soochow Univ",
"location": {
"postCode": "215006",
"settlement": "Suzhou",
"country": "China"
}
},
"email": "gdzhou@suda.edu.cn"
},
{
"first": "Junhui",
"middle": [],
"last": "Li",
"suffix": "",
"affiliation": {
"laboratory": "Jiangsu Provincial Key Lab for Computer Information Processing Technology",
"institution": "Soochow Univ",
"location": {
"postCode": "215006",
"settlement": "Suzhou",
"country": "China"
}
},
"email": "lijunhui@suda.edu.cn"
},
{
"first": "Longhua",
"middle": [],
"last": "Qian",
"suffix": "",
"affiliation": {
"laboratory": "Jiangsu Provincial Key Lab for Computer Information Processing Technology",
"institution": "Soochow Univ",
"location": {
"postCode": "215006",
"settlement": "Suzhou",
"country": "China"
}
},
"email": "qianlonghua@suda.edu.cn"
},
{
"first": "Qiaoming",
"middle": [],
"last": "Zhu",
"suffix": "",
"affiliation": {
"laboratory": "Jiangsu Provincial Key Lab for Computer Information Processing Technology",
"institution": "Soochow Univ",
"location": {
"postCode": "215006",
"settlement": "Suzhou",
"country": "China"
}
},
"email": "qmzhu@suda.edu.cn"
}
],
"year": "",
"venue": null,
"identifiers": {},
"abstract": "This paper proposes a semi-supervised learning method for relation extraction. Given a small amount of labeled data and a large amount of unlabeled data, it first bootstraps a moderate number of weighted support vectors via SVM through a co-training procedure with random feature projection and then applies a label propagation (LP) algorithm via the bootstrapped support vectors. Evaluation on the ACE RDC 2003 corpus shows that our method outperforms the normal LP algorithm via all the available labeled data without SVM bootstrapping. Moreover, our method can largely reduce the computational burden. This suggests that our proposed method can integrate the advantages of both SVM bootstrapping and label propagation.",
"pdf_parse": {
"paper_id": "I08-1005",
"_pdf_hash": "",
"abstract": [
{
"text": "This paper proposes a semi-supervised learning method for relation extraction. Given a small amount of labeled data and a large amount of unlabeled data, it first bootstraps a moderate number of weighted support vectors via SVM through a co-training procedure with random feature projection and then applies a label propagation (LP) algorithm via the bootstrapped support vectors. Evaluation on the ACE RDC 2003 corpus shows that our method outperforms the normal LP algorithm via all the available labeled data without SVM bootstrapping. Moreover, our method can largely reduce the computational burden. This suggests that our proposed method can integrate the advantages of both SVM bootstrapping and label propagation.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Abstract",
"sec_num": null
}
],
"body_text": [
{
"text": "Relation extraction is to detect and classify various predefined semantic relations between two entities from text and can be very useful in many NLP applications such as question answering, e.g. to answer the query \"Who is the president of the United States?\", and information retrieval, e.g. to expand the query \"George W. Bush\" with \"the president of the United States\" via his relationship with \"the United States\".",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "During the last decade, many methods have been proposed in relation extraction, such as supervised learning (Miller et al 2000; Zelenko et al 2003; Culota and Sorensen 2004; Zhao and Grishman 2005; Zhang et al 2006; Zhou et al 2005 Zhou et al , 2006 , semi-supervised learning (Brin 1998; Agichtein and Gravano 2000; Zhang 2004; Chen et al 2006) , and unsupervised learning (Hasegawa et al 2004; Zhang et al 2005) . Among these methods, supervised learning-based methods perform much better than the other two alternatives. However, their performance much depends on the availability of a large amount of manually labeled data and it is normally difficult to adapt an existing system to other applications and domains. On the other hand, unsupervised learning-based methods do not need the definition of relation types and the availability of manually labeled data. However, they fail to classify exact relation types between two entities and their performance is normally very low. To achieve better portability and balance between human efforts and performance, semi-supervised learning has drawn more and more attention recently in relation extraction and other NLP applications.",
"cite_spans": [
{
"start": 108,
"end": 127,
"text": "(Miller et al 2000;",
"ref_id": "BIBREF10"
},
{
"start": 128,
"end": 147,
"text": "Zelenko et al 2003;",
"ref_id": "BIBREF17"
},
{
"start": 148,
"end": 173,
"text": "Culota and Sorensen 2004;",
"ref_id": null
},
{
"start": 174,
"end": 197,
"text": "Zhao and Grishman 2005;",
"ref_id": "BIBREF22"
},
{
"start": 198,
"end": 215,
"text": "Zhang et al 2006;",
"ref_id": "BIBREF19"
},
{
"start": 216,
"end": 231,
"text": "Zhou et al 2005",
"ref_id": "BIBREF23"
},
{
"start": 232,
"end": 249,
"text": "Zhou et al , 2006",
"ref_id": "BIBREF24"
},
{
"start": 277,
"end": 288,
"text": "(Brin 1998;",
"ref_id": "BIBREF5"
},
{
"start": 289,
"end": 316,
"text": "Agichtein and Gravano 2000;",
"ref_id": "BIBREF1"
},
{
"start": 317,
"end": 328,
"text": "Zhang 2004;",
"ref_id": "BIBREF20"
},
{
"start": 329,
"end": 345,
"text": "Chen et al 2006)",
"ref_id": "BIBREF7"
},
{
"start": 374,
"end": 395,
"text": "(Hasegawa et al 2004;",
"ref_id": "BIBREF9"
},
{
"start": 396,
"end": 413,
"text": "Zhang et al 2005)",
"ref_id": "BIBREF18"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "This paper proposes a semi-supervised learning method for relation extraction. G iven a small amount of labeled data and a large amount of unlabeled data, our proposed method first bootstraps a moderate number of weighted support vectors from all the available data via SVM using a co-training procedure with random feature projection and then applies a label propagation (LP) algorithm to capture the manifold structure in both the labeled and unlabeled data via the bootstrapped support vectors. Compared with previous methods, our method can integrate the advantages of both SVM bootstrapping in learning critical instances for the labeling function and label propagation in capturing the manifold structure in both the labeled and unlabeled data to smooth the labeling function.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "The rest of this paper is as follows. In Section 2, we review related semi-supervised learning work in relation extraction. Then, the LP algorithm via bootstrapped support vectors is proposed in Section 3 while Section 4 shows the experimental results. Finally, we conclude our work in Section 5.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Introduction",
"sec_num": "1"
},
{
"text": "Generally, supervised learning is preferable to unsupervised learning due to prior knowledge in the annotated training data and better performance. However, the annotated data is usually expensive to obtain. Hence, there has been growing interest in semi-supervised learning, aiming at inducing classifiers by leveraging a small amount of labeled data and a large amount of unlabeled data. Related work in relation extraction using semi-supervised learning can be classified into two categories: bootstrapping-based (Brin 1998; Agichtein and Gravano 2000; Zhang 2004 ) and label propagation(LP)-based (Chen et al 2006) .",
"cite_spans": [
{
"start": 516,
"end": 527,
"text": "(Brin 1998;",
"ref_id": "BIBREF5"
},
{
"start": 528,
"end": 555,
"text": "Agichtein and Gravano 2000;",
"ref_id": "BIBREF1"
},
{
"start": 556,
"end": 566,
"text": "Zhang 2004",
"ref_id": "BIBREF20"
},
{
"start": 601,
"end": 618,
"text": "(Chen et al 2006)",
"ref_id": "BIBREF7"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "2"
},
{
"text": "Currently, bootstrapping-based methods dominate semi-supervised learning in relation extraction. Bootstrapping works by iteratively classifying unlabeled instances and adding confidently classified ones into labeled data using a model learned from augmented labeled data in previous iteration. Brin (1998) proposed a bootstrapping-based method on the top of a self-developed pattern matching-based classifier to exploit the duality between patterns and relations. Agichtein and Gravano (2000) shared much in common with Brin (1998) . They employed an existing pattern matching-based classifier (i.e. SNoW) instead. Zhang (2004) approached the much simpler relation classification sub-task by bootstrapping on the top of SVM. Although bootstrapping-based methods have achieved certain success, one problem is that they may not be able to well capture the manifold structure among unlabeled data.",
"cite_spans": [
{
"start": 294,
"end": 305,
"text": "Brin (1998)",
"ref_id": "BIBREF5"
},
{
"start": 464,
"end": 492,
"text": "Agichtein and Gravano (2000)",
"ref_id": "BIBREF1"
},
{
"start": 520,
"end": 531,
"text": "Brin (1998)",
"ref_id": "BIBREF5"
},
{
"start": 615,
"end": 627,
"text": "Zhang (2004)",
"ref_id": "BIBREF20"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "2"
},
{
"text": "As an alternative to the bootstrapping-based methods, Chen et al (2006) employed a LP-based method in relation extraction. Compared with bootstrapping, the LP algorithm can effectively combine labeled data with unlabeled data in the learning process by exploiting the manifold structure (e.g. the natural clustering structure) in both the labeled and unlabeled data. The rationale behind this algorithm is that the instances in highdensity areas tend to carry the same labels. The LP algorithm has also been successfully applied in other NLP applications, such as word sense disambiguation (Niu et al 2005) , text classification (Szummer and Jaakkola 2001; Blum and Chawla 2001; Belkin and Niyogi 2002; Zhu and Ghahramani 2002; Zhu et al 2003; Blum et al 2004) , and information retrieval (Yang et al 2006) . However, one problem is its computational burden, especially when a large amount of labeled and unlabeled data is taken into consideration.",
"cite_spans": [
{
"start": 54,
"end": 71,
"text": "Chen et al (2006)",
"ref_id": "BIBREF7"
},
{
"start": 590,
"end": 606,
"text": "(Niu et al 2005)",
"ref_id": "BIBREF13"
},
{
"start": 629,
"end": 656,
"text": "(Szummer and Jaakkola 2001;",
"ref_id": "BIBREF15"
},
{
"start": 657,
"end": 678,
"text": "Blum and Chawla 2001;",
"ref_id": "BIBREF3"
},
{
"start": 679,
"end": 702,
"text": "Belkin and Niyogi 2002;",
"ref_id": "BIBREF2"
},
{
"start": 703,
"end": 727,
"text": "Zhu and Ghahramani 2002;",
"ref_id": "BIBREF25"
},
{
"start": 728,
"end": 743,
"text": "Zhu et al 2003;",
"ref_id": "BIBREF26"
},
{
"start": 744,
"end": 760,
"text": "Blum et al 2004)",
"ref_id": "BIBREF4"
},
{
"start": 789,
"end": 806,
"text": "(Yang et al 2006)",
"ref_id": "BIBREF16"
}
],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "2"
},
{
"text": "In order to take the advantages of both bootstrapping and label propagation, our proposed method propagates labels via bootstrapped support vectors. On the one hand, our method can well capture the manifold structure in both the labeled and unlabeled data. On the other hand, our method can largely reduce the computational burden in the normal LP algorithm via all the available data.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Related Work",
"sec_num": "2"
},
{
"text": "The idea behind our LP algorithm via bootstrapped support vectors is that, instead of propagating labels through all the available labeled data, our method propagates labels through critical instances in both the labeled and unlabeled data. In this paper, we use SVM as the underlying classifier to bootstrap a moderate number of weighted support vectors for this purpose. This is based on an assumption that the manifold structure in both the labeled and unlabeled data can be well preserved through the critical i nstances (i.e. the weighted support vectors bootstrapped from all the available labeled and unlabeled data). The reason why we choose SVM is that it represents the state-of-theart in machine learning research and there are good implementations of the algorithm available. In particular, SVMLight (Joachims 1998) is selected as our classifier. For efficiency, we apply the one vs. others strategy, which builds K classifiers so as to separate one class from all others. Another reason is that we can adopt the weighted support vectors returned by the bootstrapped SVMs as the critical instances, via which label propagation is done.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Label ropagation via Bootstrapped Support Vectors",
"sec_num": "3"
},
{
"text": "This paper modifies the SVM bootstrapping algorithm BootProject (Zhang 2004) to bootstrap support vectors. Given a small amount of labeled data and a large amount of unlabeled data, the modified BootProject algorithm bootstraps on the top of SVM by iteratively classifying unlabeled instances and moving confidently classified ones into labeled data using a model learned from the augmented labeled data in previous iteration, until not enough unlabeled instances can be classified confidently. Figure 1 shows the modified BootProject algorithm for bootstrapping support vectors.",
"cite_spans": [
{
"start": 64,
"end": 76,
"text": "(Zhang 2004)",
"ref_id": "BIBREF20"
}
],
"ref_spans": [
{
"start": 495,
"end": 503,
"text": "Figure 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Bootstrapping Support Vectors",
"sec_num": "3.1"
},
{
"text": "L : the labeled data; U : the unlabeled data; S : the batch size (100 in our experiments); P : the number of views(feature projections); r : the number of classes (including all the relation (sub)types and the non-relation) BEGIN REPEAT FOR i = 1 to P DO Generate projected feature space i F from the original feature space F ; Project both L and U onto i F , thus gener-",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "_________________________________________ Assume:",
"sec_num": null
},
{
"text": "ate i L and i U ; Train SVM classifier ij SVM on i L for each class ) 1 ( r j r j K = ; Run ij SVM on i U for each class ) 1 ( r j r j K = END FOR",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "_________________________________________ Assume:",
"sec_num": null
},
{
"text": "Find (at most) S instances in U with the highest agreement (with threshold 70% in our experiments) and the highest average SVM-returned confidence value (with threshold 1.0 in our experiments); Move them from U to L; UNTIL not enough unlabeled instances (less than 10 in our experiments) can be confidently classified;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "_________________________________________ Assume:",
"sec_num": null
},
{
"text": "Return all the (positive and negative) support vectors included in all the latest SVM classifiers ij SVM with their collective weight (absolute alpha*y) information as the set of bootstrapped support vectors to act as the labeled data in the LP algorithm;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "_________________________________________ Assume:",
"sec_num": null
},
{
"text": "Return U (those hard cases which can not be confidently classified) to act as the unlabeled data in the LP algorithm;",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "_________________________________________ Assume:",
"sec_num": null
},
{
"text": "END _________________________________________ Figure 1 : The algorithm for bootstrapping support vectors In particular, this algorithm generates multiple overlapping \"views\" by projecting from the original feature space. In this paper, feature views with random feature projection, as proposed in Zhang (2004) , are explored. Section 4 will discuss this issue in more details. During the iterative training process, classifiers trained on the augmented labeled data using the projected views are then asked to vote on the remaining unlabeled instances and those with the highest probability of being correctly labeled are chosen to augment the labeled data.",
"cite_spans": [
{
"start": 297,
"end": 309,
"text": "Zhang (2004)",
"ref_id": "BIBREF20"
}
],
"ref_spans": [
{
"start": 46,
"end": 54,
"text": "Figure 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "_________________________________________ Assume:",
"sec_num": null
},
{
"text": "During the bootstrapping process, the support vectors included in all the trained SVM classifiers (for all the relation (sub)types and the non-relation) are bootstrapped (i.e. updated) at each iteration. When the bootstrapping process stops, all the (positive and negative) support vectors included in the SVM classifiers are returned as bootstrapped support vectors with their collective weights (absolute a*y) to act as the labeled data in the LP algorithm and all the remaining unlabeled instances (i.e. those hard cases which can not be confidently classified in the bootstrapping process) in the unlabeled data are returned to act as the unlabeled data in the LP algorithm. Through SVM bootstrapping, our LP algorithm will only depend on the critical instances (i.e. support vectors with their weight information bootstrapped from all the available labeled and unlabeled data) and those hard i nstances, instead of all the available labeled and unlabeled data.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "_________________________________________ Assume:",
"sec_num": null
},
{
"text": "In the LP algorithm (Zhu and Ghahramani 2002) , the manifold structure in data is represented as a connected graph. Given the labeled data (the above bootstrapped support vectors with their weights) and unlabeled data (the remaining hard instances in the unlabeled data after bootstrapping, including all the test instances for evaluation), the LP algorithm first represents labeled and unlabeled instances as vertices in a connected graph, then propagates the label information from any vertex to nearby vertex through weighted edges and finally infers the labels of unlabeled instances until a global stable stage is achieved. Figure 2 presents the label propagation algorithm on bootstrapped support vectors in details. Clamp the labeled data, that is, replace",
"cite_spans": [
{
"start": 20,
"end": 45,
"text": "(Zhu and Ghahramani 2002)",
"ref_id": "BIBREF25"
}
],
"ref_spans": [
{
"start": 629,
"end": 637,
"text": "Figure 2",
"ref_id": null
}
],
"eq_spans": [],
"section": "Label Propagation",
"sec_num": "3.2"
},
{
"text": "1 + t L Y with 0 L Y ; UNTIL Y converges(e.g. 1 + t L Y converges to 0 L Y );",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Label Propagation",
"sec_num": "3.2"
},
{
"text": "Assign each unlabeled instance with a label: for During the label propagation process, the label distribution of the labeled data is clamped in each loop using the weights of the bootstrapped support vectors and acts like forces to push out labels through the unlabeled data. With this push originates from the labeled data, the label boundaries will be pushed much faster along edges with larger weights and settle in gaps along those with lower weights. Ideally, we can expect that ij w across different classes should be as small as possible and ij w within the same class as big as possible. In this way, label propagation happens within the same class most likely. This algorithm has been shown to converge to a unique solution (Zhu and Ghahramani 2002) , which can be obtained without iteration in theory, and the initialization of Y U 0 (the unlabeled data) is not important since Y U 0 does not affect its estimation. However, proper initialization of Y U 0 actually helps the algorithm converge more rapidly in practice. In this paper, each row in Y U 0 is initialized to the average similarity with the labeled instances. In all our experiments, we iterate over all pairs of entity mentions occurring in the same sentence to generate potential relation instances 1 . For better evaluation, we have adopted a state-of-the-art linear kernel as similarity measurements. In our linear kernel, we apply the same feature set as described in a state-of-the-art feature-based system : word, entity type, mention level, overlap, base phrase chunking, dependency tree, parse tree and semantic information. Given above various lexical, syntactic and semantic features, multiple overlapping feature views are generated in the bootstrapping process using random feature projection (Zhang 2004) . For each feature projection in bootstrapping support vectors, a feature is randomly selected with probability p and therefore the eventually projected feature space has p*F features on average, where F is the size of the original feature space. In this paper, p and the number of different views are fine-tuned to 0.5 and 10 2 respectively using 5-fold cross validation on the training data of the ACE RDC 2003 corpus. Table 1 presents the F-measures 3 (the numbers outside the parentheses) of our algorithm using the state-of-the-art linear kernel on different sizes of the ACE RDC training data with all the remaining training data and the test data 4 as the unlabeled data on the ACE RDC 2003 corpus. In this paper, we only report the performance (averaged over 5 trials) with the percentages of 5%, 10%, 25%, 50%, 75% and 100% 5 . For example, our LP algorithm via bootstrapped (weighted) support vectors achieves the F-measure of 46.5 if using only 5% of the ACE RDC 2003 training data as the labeled data and the remaining training data and the test data in this corpus as the unlabeled data. Table 1 also compares our method with SVM and the original SVM bootstrapping algorithm BootProject(i.e. bootstrapping on the top of SVM with feature projection, as proposed in Zhang (2004) ). Finally, Table 1 compares our LP algorithm via bootstrapped (weighted by default) support vectors with other possibilities, such as the scheme via bootstrapped (un-weighted, i.e. the importance of support vectors is not differentiated) support vectors and the scheme via all the available labeled data (i.e. without SVM bootstrapping). Table 1 shows that: 1) Inclusion of unlabeled data using semisupervised learning, including the SVM bootstrapping algorithm BootProject, the normal LP algorithm via all the available labeled and unlabeled data without SVM bootstrapping, and our LP algorithms via bootstrapped (either weighted or un-weighted) support vectors, consistently improves the performance, a lthough semi-supervised learning has shown to typically decrease the performance when a lot of (enough) labeled data is available (Nigam 2001) . This may be due to the insufficiency of labeled data in the ACE RDC 2003 corpus. Actually, most of relation subtypes in the two corpora much suffer from the data sparseness problem (Zhou et al 2006) . 2) All the three LP algorithms outperform the state-of-the-art SVM classifier and the SVM bootstrapping algorithm BootProject. Especially, when a small amount of labeled data is available, the performance improvements by the LP algorithms are significant. This indicates the usefulness of the manifold structure in both labeled and unlabeled data and the powerfulness of the LP algorithm in modeling such information. 3) Our LP algorithms via bootstrapped (either weighted or un-weighted) support vectors outperforms the normal LP algorithm via all the available labeled data w/o SVM bootstrapping. For example, o ur LP algorithm via bootstrapped (weighted) support vectors outperforms the normal LP algorithm from 0.6 to 3.4 in F-measure on the ACE RDC 2003 corpus respectively when the labeled data ranges from 100% to 5%. This suggests that the manifold structure in both the labeled and unlabeled data can be well preserved via bootstrapped support vectors, especially when only a small amount of labeled data is available. This implies that weighted support vectors may represent the manifold structure (e.g. the decision boundary from where label propagation is done) better than the full set of data -an interesting result worthy more quantitative and qualitative justification in the future work. 4) Our LP algorithms via bootstrapped (weighted) support vectors perform better than LP algorithms via bootstrapped (un-weighted) support vectors by ~1.0 in F-measure on average. This suggests that bootstrapped support vectors with their weights can better represent the manifold structure in all the available labeled and unlabeled data than bootstrapped support vectors without their weights. 5) Comparison of SVM, SVM bootstrapping and label propagation with bootstrapped (weighted) support vectors shows that both bootstrapping and label propagation contribute much to the performance improvement. Table 1 also shows the increases in F-measure (the numbers inside the parentheses) if we add all the instances in the ACE RDC 2004 6 corpus into the ACE RDC 2003 corpus in consideration as unlabeled data in all the four semi-supervised learning methods. It shows that adding more unlabeled data can consistently improve the performance. For example, compared with using only 5% of the ACE RDC 2003 training data as the labeled data and the remaining training data and the test data in this corpus as the unlabeled data, including the ACE RDC 2004 corpus as the unlabeled data increases the F-measures of 1.4 and 1.0 in our LP algorithm and the normal LP algorithm respectively. Table 1 shows that the contribution grows first when the labeled data begins to increase and reaches a maximum of ~2.0 in F-measure at a certain point.",
"cite_spans": [
{
"start": 733,
"end": 758,
"text": "(Zhu and Ghahramani 2002)",
"ref_id": "BIBREF25"
},
{
"start": 1778,
"end": 1790,
"text": "(Zhang 2004)",
"ref_id": "BIBREF20"
},
{
"start": 3068,
"end": 3080,
"text": "Zhang (2004)",
"ref_id": "BIBREF20"
},
{
"start": 3917,
"end": 3929,
"text": "(Nigam 2001)",
"ref_id": "BIBREF12"
},
{
"start": 4113,
"end": 4130,
"text": "(Zhou et al 2006)",
"ref_id": "BIBREF24"
}
],
"ref_spans": [
{
"start": 2212,
"end": 2219,
"text": "Table 1",
"ref_id": null
},
{
"start": 2892,
"end": 2899,
"text": "Table 1",
"ref_id": null
},
{
"start": 3093,
"end": 3100,
"text": "Table 1",
"ref_id": null
},
{
"start": 3420,
"end": 3427,
"text": "Table 1",
"ref_id": null
},
{
"start": 6040,
"end": 6047,
"text": "Table 1",
"ref_id": null
},
{
"start": 6718,
"end": 6725,
"text": "Table 1",
"ref_id": null
}
],
"eq_spans": [],
"section": "Label Propagation",
"sec_num": "3.2"
},
{
"text": ") ( n i l x i \u2264 p",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Label Propagation",
"sec_num": "3.2"
},
{
"text": "Finally, it is found in our experiments that critical and hard instances normally occupy only 15~20% (~18% on average) of all the available labeled and unlabeled data. This suggests that, through bootstrapped support vectors, our LP algo-rithm can largely reduce the computational burden since it only depends on the critical instances (i.e. bootstrapped support vectors with their weights) and those hard instances.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Experimental Results",
"sec_num": "4.2"
},
{
"text": "This paper proposes a new effective and efficient semi-supervised learning method in relation extraction. First, a moderate number of weighted support vectors are bootstrapped from all the available labeled and unlabeled data via SVM through a co-training procedure with feature projection. Here, a random feature projection technique is used to generate multiple overlapping feature views in bootstrapping using a state-of-the-art linear kernel. Then, a LP algorithm is applied to propagate labels via the bootstrapped support vectors, which, together with those hard unlabeled instances and the test instances, are represented as vertices in a connected graph. During the classification process, the label information is propagated from any vertex to nearby vertex through weighted edges and finally the labels of unlabeled instances are inferred until a global stable stage is achieved. In this way, the manifold structure in both the labeled and unlabeled data can be well captured by label propagation via bootstrapped support vectors. Evaluation on the ACE RDC 2004 corpus suggests that our LP algorithm via bootstrapped support vectors can take the advantages of both SVM bootstrapping and label propagation.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "5"
},
{
"text": "For the f uture work, we will systematically evaluate our proposed method on more corpora and explore better metrics of measuring the similarity between two instances.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Conclusion",
"sec_num": "5"
},
{
"text": "In this paper, we only measure the performance of relation extraction on \"true\" mentions with \"true\" chaining of co-reference (i.e. as annotated by the corpus annotators) in the ACE corpora. We also explicitly model the argument order of the two mentions involved and only model explicit relations because of poor inter-annotator agreement in the annotation of implicit relations and their limited number.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "This suggests that the modified BootProject algorithm in the bootstrapping phase outperforms the SelfBoot algorithm (with p=1.0 and m=1) which uses all the features as the only view. In the related NLP literature, co-training has also shown to typically outperform self-bootstrapping.3 Our experimentation also shows that most of performance improvement with either bootstrapping or label propagation comes from gain in recall. Due to space limitation, this p aper only reports the overall Fmeasure.4 In our label propagation algorithm via bootstrapped support vectors, the test data is only included in the second phase (i.e. the label propagation phase) and not used in the first phase (i.e. bootstrapping support vectors). This is to fairly compare different semisupervised learning methods.5 We have tried less percentage than 5%. However, our experiments show that using much less data will suffer from performance un-stability. Therefore, we only report the performance with percentage not less than 5%.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
},
{
"text": "Compared with the ACE RDC 2003 task, the ACE RDC 2004 task defines two more entity types, i.e. weapon and vehicle, much more entity subtypes, and different 7 relation types and 23 subtypes between 7 entity types. The ACE RDC 2004 corpus from LDC contains 451 documents and 5702 relation instances.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "",
"sec_num": null
}
],
"back_matter": [
{
"text": "This research is supported by Project 60673041 under the National Natural Science Foundation of China and Project 2006AA01Z147 under the \"863\" National High-Tech Research and Development of China.",
"cite_spans": [],
"ref_spans": [],
"eq_spans": [],
"section": "Acknowledgement",
"sec_num": null
}
],
"bib_entries": {
"BIBREF0": {
"ref_id": "b0",
"title": "Automatic Content Extraction",
"authors": [],
"year": 2000,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "ACE. (2000-2005). Automatic Content Extraction. http://www.ldc.upenn.edu/Projects/ACE/",
"links": null
},
"BIBREF1": {
"ref_id": "b1",
"title": "Snowball: Extracting relations from large plain-text collections",
"authors": [
{
"first": "E",
"middle": [],
"last": "Agichtein",
"suffix": ""
},
{
"first": "L",
"middle": [],
"last": "Gravano",
"suffix": ""
}
],
"year": 2000,
"venue": "Proceedings of the 5 th ACM International Conference on Digital Libraries",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Agichtein E. and Gravano L. (2000). Snowball: Extracting relations from large plain-text collec- tions. Proceedings of the 5 th ACM International Conference on Digital Libraries (ACMDL'2000).",
"links": null
},
"BIBREF2": {
"ref_id": "b2",
"title": "Using Manifold Structure for Partially Labeled Classification",
"authors": [
{
"first": "M",
"middle": [],
"last": "Belkin",
"suffix": ""
},
{
"first": "P",
"middle": [],
"last": "Niyogi",
"suffix": ""
}
],
"year": 2002,
"venue": "NIPS",
"volume": "15",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Belkin, M. and Niyogi, P. (2002). Using Manifold Structure for Partially Labeled Classification. NIPS 15.",
"links": null
},
"BIBREF3": {
"ref_id": "b3",
"title": "Learning from labeled and unlabeled data using graph mincuts",
"authors": [
{
"first": "A",
"middle": [],
"last": "Blum",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Chawla",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Blum A. and Chawla S. (2001). Learning from la- beled and unlabeled data using graph mincuts. ICML'2001.",
"links": null
},
"BIBREF4": {
"ref_id": "b4",
"title": "Semi-supervised learning using randomized mincuts",
"authors": [
{
"first": "A",
"middle": [],
"last": "Blum",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Lafferty",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Reddy",
"suffix": ""
}
],
"year": 2004,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Blum A., Lafferty J., Rwebangira R and Reddy R. (2004). Semi-supervised learning using random- ized mincuts. ICML'2004.",
"links": null
},
"BIBREF5": {
"ref_id": "b5",
"title": "Extracting patterns and relations from world wide web",
"authors": [
{
"first": "S",
"middle": [],
"last": "Brin",
"suffix": ""
}
],
"year": 1998,
"venue": "Proceedings of WebDB Workshop at 6 th International Conference on Extending Database Technology",
"volume": "",
"issue": "",
"pages": "172--183",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Brin S. (1998). Extracting patterns and relations from world wide web. Proceedings of WebDB Workshop at 6 th International Conference on Extending Database Technology:172-183.",
"links": null
},
"BIBREF6": {
"ref_id": "b6",
"title": "Immediate-head Parsing for Language Models",
"authors": [
{
"first": "E",
"middle": [],
"last": "Charniak",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "",
"issue": "",
"pages": "129--137",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Charniak E. (2001). Immediate-head Parsing for Language Models. ACL'2001: 129-137. Tou- louse, France",
"links": null
},
"BIBREF7": {
"ref_id": "b7",
"title": "Relation extraction using label propagation based semi-supervised learning",
"authors": [
{
"first": "J",
"middle": [
"X"
],
"last": "Chen",
"suffix": ""
},
{
"first": "D",
"middle": [
"H"
],
"last": "Ji",
"suffix": ""
},
{
"first": "C",
"middle": [
"L"
],
"last": "Tan",
"suffix": ""
},
{
"first": "Z",
"middle": [
"Y"
],
"last": "Niu",
"suffix": ""
}
],
"year": 2006,
"venue": "COLING-ACL",
"volume": "",
"issue": "",
"pages": "129--136",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Chen J.X., Ji D.H., Tan C.L. and Niu Z.Y. (2006). Relation extraction using label propagation based semi-supervised learning. COLING- ACL'2006: 129-136. July 2006. Sydney, Austra- lia.",
"links": null
},
"BIBREF8": {
"ref_id": "b8",
"title": "Dependency tree kernels for relation extraction",
"authors": [
{
"first": "A",
"middle": [],
"last": "Culotta",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Sorensen",
"suffix": ""
}
],
"year": 2004,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Culotta A. and Sorensen J. (2004). Dependency tree kernels for relation extraction. ACL'2004. 423-429. 21-26 July 2004. Barcelona, Spain.",
"links": null
},
"BIBREF9": {
"ref_id": "b9",
"title": "Discovering relations among named entities form large corpora",
"authors": [
{
"first": "T",
"middle": [],
"last": "Hasegawa",
"suffix": ""
},
{
"first": "S",
"middle": [],
"last": "Sekine",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Grishman",
"suffix": ""
}
],
"year": 2004,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Hasegawa T., Sekine S. and Grishman R. (2004). Discovering relations among named entities form large corpora. ACL'2004. Barcelona, Spain.",
"links": null
},
"BIBREF10": {
"ref_id": "b10",
"title": "A novel use of statistical parsing to extract information from text",
"authors": [
{
"first": "S",
"middle": [],
"last": "Miller",
"suffix": ""
},
{
"first": "H",
"middle": [],
"last": "Fox",
"suffix": ""
},
{
"first": "L",
"middle": [],
"last": "Ramshaw",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Weischedel",
"suffix": ""
}
],
"year": 2000,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Miller S., Fox H., Ramshaw L. and Weischedel R. (2000). A novel use of statistical parsing to ex- tract information from text. ANLP'2000. 226- 233. 29 April -4 May 2000, Seattle, USA",
"links": null
},
"BIBREF11": {
"ref_id": "b11",
"title": "A study on convolution kernels for shallow semantic parsing",
"authors": [
{
"first": "A",
"middle": [],
"last": "Moschitti",
"suffix": ""
}
],
"year": 2004,
"venue": "",
"volume": "",
"issue": "",
"pages": "335--342",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Moschitti A. (2004). A study on convolution ker- nels for shallow semantic parsing. ACL'2004:335-342.",
"links": null
},
"BIBREF12": {
"ref_id": "b12",
"title": "Using unlabeled data to improve text classification",
"authors": [
{
"first": "K",
"middle": [
"P"
],
"last": "Nigam",
"suffix": ""
}
],
"year": 2001,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Nigam K.P. (2001). Using unlabeled data to im- prove text classification. Technical Report CMU-CS-01-126.",
"links": null
},
"BIBREF13": {
"ref_id": "b13",
"title": "Word Sense Disambiguation Using Label Propagation Based Semi-supervised Learning",
"authors": [
{
"first": "Z",
"middle": [
"Y"
],
"last": "Niu",
"suffix": ""
},
{
"first": "D",
"middle": [
"H"
],
"last": "Ji",
"suffix": ""
},
{
"first": "C",
"middle": [
"L"
],
"last": "Tan",
"suffix": ""
}
],
"year": 2005,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Niu Z.Y., Ji D.H., and Tan C.L. (2005). Word Sense Disambiguation Using Label Propagation Based Semi-supervised Learning.",
"links": null
},
"BIBREF15": {
"ref_id": "b15",
"title": "Partially Labeled Classification with Markov Random Walks",
"authors": [
{
"first": "M",
"middle": [],
"last": "Szummer",
"suffix": ""
},
{
"first": "T",
"middle": [],
"last": "Jaakkola",
"suffix": ""
}
],
"year": 2001,
"venue": "NIPS",
"volume": "14",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Szummer, M., & Jaakkola, T. (2001). Partially La- beled Classification with Markov Random Walks. NIPS 14.",
"links": null
},
"BIBREF16": {
"ref_id": "b16",
"title": "Document Re-ranking using cluster validation and label propagation",
"authors": [
{
"first": "L",
"middle": [
"P"
],
"last": "Yang",
"suffix": ""
},
{
"first": "D",
"middle": [
"H"
],
"last": "Ji",
"suffix": ""
},
{
"first": "G",
"middle": [
"D"
],
"last": "Zhou",
"suffix": ""
},
{
"first": "Y",
"middle": [],
"last": "Nie",
"suffix": ""
}
],
"year": 2006,
"venue": "CIKM'",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Yang L.P., Ji D.H., Zhou G.D. and Nie Y. (2006). Document Re-ranking using cluster validation and label propagation. CIKM'2006. 5-11 Nov 2006. Arlington, Virginia, USA.",
"links": null
},
"BIBREF17": {
"ref_id": "b17",
"title": "Kernel methods for relation extraction",
"authors": [
{
"first": "D",
"middle": [],
"last": "Zelenko",
"suffix": ""
},
{
"first": "C",
"middle": [],
"last": "Aone",
"suffix": ""
},
{
"first": "",
"middle": [],
"last": "Richardella",
"suffix": ""
}
],
"year": 2003,
"venue": "Journal of Machine Learning Research",
"volume": "3",
"issue": "",
"pages": "1083--1106",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zelenko D., Aone C. and Richardella. (2003). Ker- nel methods for relation extraction. Journal of Machine Learning Research. 3(Feb):1083-1106.",
"links": null
},
"BIBREF18": {
"ref_id": "b18",
"title": "Discovering Relations from a Large Raw Corpus Using Tree Similarity-based Clustering",
"authors": [
{
"first": "M",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Su",
"suffix": ""
},
{
"first": "D",
"middle": [
"M"
],
"last": "Wang",
"suffix": ""
},
{
"first": "G",
"middle": [
"D"
],
"last": "Zhou",
"suffix": ""
},
{
"first": "C",
"middle": [
"L"
],
"last": "Tan",
"suffix": ""
}
],
"year": 2005,
"venue": "Lecture Notes in Artificial Intelligence",
"volume": "",
"issue": "",
"pages": "378--389",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhang M., Su J., Wang D.M., Zhou G.D. and Tan C.L. (2005). Discovering Relations from a Large Raw Corpus Using Tree Similarity-based Clustering, IJCNLP'2005, Lecture Notes in Arti- ficial Intelligence (LNAI 3651). 378-389.",
"links": null
},
"BIBREF19": {
"ref_id": "b19",
"title": "A Composite Kernel to Extract Relations between Entities with both Flat and Structured Features",
"authors": [
{
"first": "M",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Su",
"suffix": ""
},
{
"first": "G",
"middle": [
"D"
],
"last": "Zhou",
"suffix": ""
}
],
"year": 2006,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhang M., Zhang J., Su J. and Zhou G.D. (2006). A Composite Kernel to Extract Relations be- tween Entities with both Flat and Structured Features. COLING-ACL-2006: 825-832. Sydney, Australia",
"links": null
},
"BIBREF20": {
"ref_id": "b20",
"title": "Weakly supervised relation classification for information extraction",
"authors": [
{
"first": "Z",
"middle": [],
"last": "Zhang",
"suffix": ""
}
],
"year": 2004,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhang Z. (2004). Weakly supervised relation clas- sification for information extraction.",
"links": null
},
"BIBREF22": {
"ref_id": "b22",
"title": "Extracting relations with integrated information using kernel methods",
"authors": [
{
"first": "S",
"middle": [
"B"
],
"last": "Zhao",
"suffix": ""
},
{
"first": "R",
"middle": [],
"last": "Grishman",
"suffix": ""
}
],
"year": 2005,
"venue": "",
"volume": "",
"issue": "",
"pages": "25--30",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhao S.B. and Grishman R. (2005). Extracting re- lations with integrated information using kernel methods. ACL'2005: 419-426. Univ of Michi- gan-Ann Arbor, USA, 25-30 June 2005.",
"links": null
},
"BIBREF23": {
"ref_id": "b23",
"title": "Exploring various knowledge in relation extraction",
"authors": [
{
"first": "G",
"middle": [
"D"
],
"last": "Zhou",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Su",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Zhang",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Zhang",
"suffix": ""
}
],
"year": 2005,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhou G.D., Su J. Zhang J. and Zhang M. (2005). Exploring various knowledge in relation extrac- tion. ACL'2005. 427-434. 25-30 June, Ann Ar- bor, Michgan, USA.",
"links": null
},
"BIBREF24": {
"ref_id": "b24",
"title": "Modeling commonality among related classes in relation extraction",
"authors": [
{
"first": "G",
"middle": [
"D"
],
"last": "Zhou",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Su",
"suffix": ""
},
{
"first": "M",
"middle": [],
"last": "Zhang",
"suffix": ""
}
],
"year": 2006,
"venue": "COLING-ACL",
"volume": "",
"issue": "",
"pages": "121--128",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhou G.D., Su J. and Zhang M. (2006). Modeling commonality among related classes in relation extraction, COLING-ACL'2006: 121-128. Syd- ney, Australia.",
"links": null
},
"BIBREF25": {
"ref_id": "b25",
"title": "Learning from Labeled and Unlabeled Data with Label Propagation",
"authors": [
{
"first": "X",
"middle": [],
"last": "Zhu",
"suffix": ""
},
{
"first": "Z",
"middle": [],
"last": "Ghahramani",
"suffix": ""
}
],
"year": 2002,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhu, X. and Ghahramani, Z. (2002). Learning from Labeled and Unlabeled Data with Label Propagation. CMU CALD Technical Report. CMU-CALD-02-107.",
"links": null
},
"BIBREF26": {
"ref_id": "b26",
"title": "Semi-Supervised Learning Using Gaussian Fields and Harmonic Functions",
"authors": [
{
"first": "X",
"middle": [],
"last": "Zhu",
"suffix": ""
},
{
"first": "Z",
"middle": [],
"last": "Ghahramani",
"suffix": ""
},
{
"first": "J",
"middle": [],
"last": "Lafferty",
"suffix": ""
}
],
"year": 2003,
"venue": "",
"volume": "",
"issue": "",
"pages": "",
"other_ids": {},
"num": null,
"urls": [],
"raw_text": "Zhu, X., Ghahramani, Z. and Lafferty, J. (2003). Semi-Supervised Learning Using Gaussian Fields and Harmonic Functions. ICML'2003.",
"links": null
}
},
"ref_entries": {
"FIGREF1": {
"text": "vertex corresponds to an instance, and the edge between any two instances i x and j x is weighted by ij w to measure their similarity. In principle, larger edge weights allow labels to travel through easier. Thus the closer the instances are, the more likely they have similar labels. The algorithm first calculates the weight ij w probability interpretation of the labeling matrix Y .",
"num": null,
"type_str": "figure",
"uris": null
},
"TABREF1": {
"type_str": "table",
"num": null,
"html": null,
"text": "Comparison of different methods using a state-of-the-art linear kernel on the ACE RDC 2003 corpus (The numbers inside the parentheses indicate the increases in F-measure if we add the ACE RDC 2004 corpus as the unlabeled data)",
"content": "<table><tr><td>Method</td><td>LP via bootstrapped (weighted) SVs</td><td>LP via bootstrapped (un-weighted) SVs</td><td>LP w/o SVM bootstrapping</td><td>SVM</td><td>(BootProject) SVM Bootstrapping</td></tr><tr><td>5%</td><td>46.5 (+1.4)</td><td>44.5 (+1.7)</td><td>43.1 (+1.0)</td><td>35.4 (-)</td><td>40.6 (+0.9)</td></tr><tr><td>10%</td><td>48.6 (+1.7)</td><td>46.5 (+2.1)</td><td>45.2 (+1.5)</td><td>38.6 (-)</td><td>43.1 (+1.4)</td></tr><tr><td>25%</td><td>51.7 (+1.9)</td><td>50.4 (+2.3)</td><td>49.6 (+1.8)</td><td>43.9 (-)</td><td>47.8 (+1.7)</td></tr><tr><td>50%</td><td>53.6 (+1.8)</td><td>52.6 (+2.2)</td><td>52.1 (+1.7)</td><td>47.2 (-)</td><td>50.5 (+1.6)</td></tr><tr><td>75%</td><td>55.2 (+1.3)</td><td>54.5 (+1.8)</td><td>54.2 (+1.2)</td><td>53.1 (-)</td><td>53.9 (+1.2)</td></tr><tr><td>100%</td><td>56.2 (+1.0)</td><td>55.8 (+1.3)</td><td>55.6 (+0.8)</td><td>55.5 (-)</td><td>55.8 (+0.7)</td></tr><tr><td colspan=\"2\">Table 1: 4.1 Experimental Setting</td><td/><td/><td/><td/></tr><tr><td colspan=\"3\">In the ACE RDC 2003 corpus, the training data</td><td/><td/><td/></tr><tr><td colspan=\"3\">consists of 674 annotated text documents (~300k</td><td/><td/><td/></tr><tr><td colspan=\"3\">words) and 9683 instances of relations. During</td><td/><td/><td/></tr><tr><td colspan=\"3\">development, 155 of 674 documents in the training</td><td/><td/><td/></tr><tr><td colspan=\"3\">set are set aside for fine-tuning. The test set is held</td><td/><td/><td/></tr><tr><td colspan=\"3\">out only for final evaluation. It consists of 97</td><td/><td/><td/></tr><tr><td colspan=\"3\">documents (~50k words) and 1386 instances of</td><td/><td/><td/></tr><tr><td colspan=\"3\">relations. The ACE RDC 2003 task defines 5 rela-</td><td/><td/><td/></tr><tr><td colspan=\"3\">tion types and 24 subtypes between 5 entity types,</td><td/><td/><td/></tr><tr><td colspan=\"3\">i.e. person, organization, location, facility and GPE.</td><td/><td/><td/></tr><tr><td colspan=\"3\">All the evaluations are measured on the 24 sub-</td><td/><td/><td/></tr><tr><td colspan=\"3\">types including relation identification and classifi-</td><td/><td/><td/></tr><tr><td>cation.</td><td/><td/><td/><td/><td/></tr><tr><td/><td/><td/><td colspan=\"3\">This paper uses the ACE RDC 2003 corpus pro-</td></tr><tr><td/><td/><td/><td colspan=\"3\">vided by LDC for evaluation. This corpus is gath-</td></tr><tr><td/><td/><td/><td colspan=\"3\">ered f rom various newspapers, newswires and</td></tr><tr><td/><td/><td/><td>broadcasts.</td><td/><td/></tr></table>"
}
}
}
} |